To appear Environmental Science and Technology, accepted...
Transcript of To appear Environmental Science and Technology, accepted...
To appear Environmental Science and Technology, accepted April 2011
1
REMANUFACTURING AND ENERGY SAVINGS
Timothy G. Gutowski∗a, Sahil Sahnib, Avid Boustania, Stephen C. Gravesa,c
a. Department of Mechanical Engineering, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
b. Department of Materials Science and Engineering, Massachusetts Institute of
Technology, Cambridge, Massachusetts 02139
c. Sloan School of Management, Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139
Abstract
Remanufactured products that can substitute for new products are generally claimed to save energy. These claims are made from studies that look mainly at the differences in materials production and manufacturing. However, when the use phase is included, the situation can change radically. In this paper, 25 case studies for eight different product categories were studied, including: 1) furniture, 2) clothing, 3) computers, 4) electric motors, 5) tires, 6) appliances, 7) engines and 8) toner cartridges. For most of these products, the use phase energy dominates that for materials production and manufacturing combined. As a result, small changes in use phase efficiency can overwhelm the claimed savings from materials production and manufacturing. These use phase energy changes are primarily due to efficiency improvements in new products, and efficiency degradation in remanufactured products. For those products with no, or an unchanging use phase energy requirement, remanufacturing can save energy. For the 25 cases, we found that 8 cases clearly saved energy, 6 did not, and 11 were too close to call. In some cases we could examine how the energy savings potential of remanufacturing has changed over time. Specifically, during times of significant improvements in energy efficiency, remanufacturing would often not save energy. A general design trend seems to be to add power to a previously unpowered product, and then to improve on the energy efficiency of the product over time. These trends tend to undermine the energy savings potential of remanufacturing.
1. Introduction to Remanufacturing
Remanufacturing is generally seen as the most environmentally friendly of “end of life”
treatments for a retired product. If the remanufactured product can be considered a
substitute for a new product, then a credit is usually claimed for the avoided resource use
∗ Corresponding Author Phone: (617) 253-2034; fax: (617) 253-1556; e-mail: [email protected]
To appear Environmental Science and Technology, accepted April 2011
2
and emissions associated with the new product production. The biggest savings is
generally from the avoided new materials production, but the difference between new
manufacturing and remanufacturing can also be significant. At the same time,
remanufactured products generally sell for about 50-80% of the new product. Hence
remanufacturing can be seen as a win-win; it saves money (for the consumer) and it saves
the environment.
In the United States, remanufacturing is at least a $50 billion industry with direct
employment of about 480,000 in 73,000 firms (1). Remanufactured products include
automotive and aircraft parts, compressors and electrical motors, office furniture, tires,
toner cartridges, office equipment, machine tools, cameras and still others (1). One of the
primary requirements for remanufacturing is that the retired products have significant
residual value at the end of life. The second is that the remanufacturing firm can
effectively capture the retired product. And the third is that the product can be restored to
like-new condition (in terms of product function) with only a modest investment. In terms
of number of remanufacturing plants, the largest remanufacturing categories in the US
are tires, followed by motors and generators and motor vehicle parts (2).
The fact that a product can have significant residual value at its end of life can present a
dilemma for the original equipment manufacturer (OEM). For example, if the OEM
decides to not remanufacture its own products, then it might find itself competing with its
own products remanufactured by another firm. To avoid being placed in this situation, an
OEM might employ a variety of strategies to defeat “third party” remanufacturing. These
strategies might include making spent products inoperable, rapid (minor) design changes,
using a “prebate” system, and buying back the spent products. All of these strategies
have been employed by various printer OEMs with varying success in an effort to protect
their ink cartridge business. For example, the prebate system employed by Lexmark,
attempts to enter into a contractual agreement with the buyer to return or throw away the
spent ink cartridge in exchange for a discount. However, the U.S. District Court of
Kentucky barred this practice recently citing a U.S. Supreme Court 2008 decision in
To appear Environmental Science and Technology, accepted April 2011
3
Quanta Vs. LG Electronics (3), interpreting it as an attempt to avoid the patent exhaustion
doctrine.
An alternative position is to embrace remanufacturing and to make it part of the OEM’s
business strategy. A variety of firms have done this, particularly for truck tires and heavy
equipment (e.g. Caterpillar, Cummins, Goodyear, Michelin). This strategy can build a
strong long term relationship with customers. As a general method for supplying
products to customers however, remanufacturing presents some challenges. One
challenge is to match supply and demand. The early steps in remanufacturing, which
consist of recovering the spent product (sometimes called “the core”), cleaning it and
testing it, all represent an investment. To capture the value of that investment and to
guard against fluctuations in core supply, a remanufacturer may have to maintain a large
inventory of cleaned and tested cores. A second challenge is that remanufacturing is
labor intensive. The condition and variety of incoming cores can vary significantly. This
means that remanufacturing must be flexible. Hence two conditions that favor
remanufacturing are: 1) a relatively low wage, skilled labor market, and 2) modest
inventory storage costs. In addition to this, the remanufacturer will need to have an
effective way to recover spent cores.
2. Research Questions
The aim of this paper is to test the hypothesis ‘remanufacturing of products saves
energy,’ as popularly claimed. The research questions that motivated this study, were: 1)
how big is the energy savings potential of remanufacturing, with a particular interest in
identifying the products that represent the best opportunities for energy savings, and 2)
how could this energy savings potential be expanded, both in terms of remanufacturing
more of the usual category of products, and to expand to new product categories.
To address these issues, we studied 8 different product categories: 1) furniture, 2)
clothing, 3) computers, 4) electric motors, 5) tires, 6) appliances, 7) engines, and 8) toner
To appear Environmental Science and Technology, accepted April 2011
4
cartridges, many with very high remanufacturing potential in the United States. The
analysis was framed in terms of a product replacement decision for a consumer in the
U.S. That is, we pose a scenario in which a consumer intends to replace a product and
we examine the normative question: to save energy, should the consumer acquire a
remanufactured version of the retired product, or should the consumer buy new? The
question is answered by using a life cycle energy analysis for the two product options.
The analysis includes the energy requirements for materials production, manufacturing
and the product use phase. We perform a sensitivity analysis to consider elements that
were not included in the analysis, as well as parameter variation. Variations on system
boundaries and elements beyond the life cycle of a single product are discussed at the end
of the paper
3. Life Cycle Energy Analysis of Products
The life cycle energy analysis of products is now a well-established field of study. Many
studies have already been performed, many software programs are available to help in
this analysis, and international standards exist to guide the practitioner. In this study, we
take advantage of the analyses by others for products that fit into the general categories
for remanufacturable products. In order to double check these studies, and to resolve
differences between multiple studies for similar products, we developed a life cycle
energy estimation tool for materials production and manufacturing (4). The tool only
requires a bill of materials (BOM) for the product and uses well known estimates both for
the embodied energy in materials (5) (6), and for the energy requirements for various
manufacturing processes (7) (4). Comparisons between the life cycle energy results from
others and our model helped validate the accuracy of the data used in this study. The
Supporting Information (S.I.) provides the model used to conduct the analyses.
Others have also addressed related questions in the literature such as in studies on the
remanufacturing of specific products, optimum product replacement strategies, and
product leasing (8) (9) (10) (11) (12) (13) (14) (15) (16). An important and generally
well-known result from product life cycle studies is that for most products the energy
To appear Environmental Science and Technology, accepted April 2011
5
requirement for materials production dominates the energy requirements for
manufacturing. In addition, observations from remanufacturing studies show that most of
the original materials in the remanufactured product are saved, and the energy required
for remanufacturing is almost always much less than that required for the original
manufacturing (1) (17) (18) (19).
A second common observation from life cycle analysis (LCA) studies for “powered”
products, which require an energy source, is that it is very common for the use phase to
dominate energy use. That is, the energy requirements of the use phase can exceed the
combined requirements of both materials production and manufacturing. As a result, as
will be seen, even small changes in use phase energy can produce significantly different
outcomes.
Based on these observations, for some products in this study we chose to ignore the
energy requirements for remanufacturing. This, of course, will bias the results slightly in
favor of remanufacturing; however as will be seen, the effect is generally negligible.
Furthermore, this simplification opens up the interpretation of these results to include
several other categories of product restoration such as repairing, refurbishing and even re-
selling if the product is still in like-new condition.
4. Research Results
We start with two representative cases, refrigerators and heavy-duty truck tires, which
illustrate the methodology, and point out special issues that can arise. The analysis that
follows considers a product retired in year X after a first lifetime of L years. The
comparison is between a like-new, but remanufactured product of model year (X – L)
versus a new product of model year X. For example, for a computer with a purchase-to-
purchase life of 4 years (L), we compare a new device of 2005 (X) with a used one of
2001 (X-L). Because of the magnitude of the use phase energy for powered products, we
will pay particular attention to changes in usage patterns and changes in energy
performance in the U.S. In addition, we normalize the analysis to account for
To appear Environmental Science and Technology, accepted April 2011
6
improvements to the product that can be captured in the functional unit, e.g. larger
refrigerators and longer lasting tires.
4.1 Refrigerators
Consider the case of a refrigerator that breaks down in year X after a first life of L years
of service because of a failed compressor. All other functions for the refrigerator perform
at their like-new level corresponding to their original model year (X – L). (This of course
is an optimistic statement favoring remanufacturing). The options considered in this
analysis are to replace the failed compressor with one that has been remanufactured and
use the refrigerator for another L years, or to buy new. The analysis calculates the life
cycle energy requirements per cubic meter of cooled space for the materials,
manufacturing and use phases. In this case, we assume that the materials and
manufacturing energy requirements to remanufacture the product are zero. The analysis
is performed for four cases corresponding to new model years in 1966, 1980, 1994 and
2008. The key assumptions used in this study pertain to the use phase energy
requirements, which have already been well documented for medium sized residential
refrigerators used in the U.S. over the time period 1947-2008. See in particular
Rosenfeld 2003 (20) and AHAM 2008 (21); also Kim 2005 (11) gives a good review of
this topic. The data show that over this time period, the electricity requirements per unit
go from about 350 kWh/yr in 1947, to a peak of about 1850 kWh/yr in 1974, to about 450
kWh/yr in 2008. The early rise is due to added features (e.g. defrosting, larger freezer)
and increases in size, while the decrease is due to energy efficiency mandates first in
California in 1974 and later at the federal level. During this time period refrigerators
grew from about 0.23m3 cooled volume in 1947 to 0.61m3 in 2008. DOE sets the typical
service lifetime for a refrigerator in the range of 10 -16 years. In this study we assume a
lifetime of L = 14 years (also used by (22), (23)). Using these values means that a 2008
new refrigerator will use 6300 kWh electricity or (using a grid efficiency of 1/3) about 68
GJ of primary energy over its lifetime. To estimate the primary energy requirements for
To appear Environmental Science and Technology, accepted April 2011
7
the materials and manufacturing of a modern refrigerator we reviewed the studies of
others ((24), (11) (25), (26)) and using the bill of materials provided by (11) applied our
model (4). The results give a range of 4,442 to 6,847 MJ for a late 1990s era model with
0.59m3 of cooled space. Again, to be conservative (in favor of remanufacturing) we used
the higher value of about 6.9GJ and assume that it is applicable to 2008, which is argued
below. Comparing this with 68 GJ one sees that the use phase energy is larger by about a
factor of 10. In the 1974 case (the peak year for energy use per refrigerator) the ratio is
over 40, and in the 1952 case (the remanufactured model year for the 1966 new
comparison) the ratio is about 9. The upshot of this is that materials and manufacturing
energy play a relatively small role in the life cycle energy requirements of a refrigerator.
During the 56 years examined here, it is true that materials and manufacturing energy
would have changed, but we argue that these changes were probably not much different
than the original range given earlier (4.4 to 6.9GJ). This case can be made based upon
the observation that among the various changes, the two most important probably
cancelled each other out. That is, among design changes one would expect more
optimized use of materials (e.g. thinner sections), materials substitution (mostly plastics
for ferrous metals) and most importantly larger size (a factor of 2.1 from 1952 to 2008).
The size change however would probably be offset by increased efficiency in materials
production. Trends given by Smil 2009 (5), Chapman and Roberts 1983 (27) and
Dahmus and Gutowski 2011 (28) suggest that a reasonable estimate for efficiency
improvements for ferrous metal production (the dominant material in refrigerators) would
be about 1.5% per year. At this rate, over 56 years, yields a factor of improvement of
about 2.3. Overall, then it appears that increases in energy due to 1) the substitution of
plastics for ferrous metals and, 2) an increase in size, would be offset by improvements in
design, and materials production efficiency. Improvement in parts manufacturing would
also have increases (injection molding and thermoforming of plastics vs. sheet forming)
and decreases due to efficiency improvements, but overall would be insignificant.
Putting this all together we show in Figure 1 the new-product life cycle energy plotted on
the Y axis, and the remanufactured-product life cycle energy plotted on the X axis.
Points above the dividing line favor remanufacturing, while points below the line favor
To appear Environmental Science and Technology, accepted April 2011
8
buying new. Following the points around the figure shows that in the early years when
use energy was increasing, remanufacturing is favored; however, after 1974
improvements in use phase efficiency favored buying new. In the inset one can see the
resulting life cycle energy for the four new model years (1966, 1980, 1994, and 2008).
The life cycle energy for the remanufacturing case is essentially the earlier model year
(e.g. 1966 for the 1980 comparison) minus the materials and manufacturing contribution.
The figure clearly shows how small the materials and manufacturing energy is compared
to the use phase energy.
4.2 Heavy Duty Truck Tires
For a second illustrative example, consider the decision to replace a spent truck tire with
a new or an “equivalent” retreaded tire. Retreading truck tires is a big business in the
United States. According to Michelin about 44 % of all replacement tires are retreaded
(29). From a life cycle analysis perspective there are several important differences
between this example and the previous example for refrigerators. The first difference is
that we found fewer life cycle studies in the literature for tires and far more variation in
the available data, in particular concerning rolling resistance and the tire use phase.
Secondly, the life span of a truck tire is far shorter than that of a home refrigerator.
Driving at 50 mph for 8 hours a day, 5 days a week for 50 weeks adds up to 100,000
miles in one year, equal to the tire lifetime. Hence historical changes in use phase
efficiency are far less important than the technology options a decision maker has when
he or she goes to replace a tire. Additionally, retreading adds significant new material to
the old casing, and is in itself an energy intensive process. As a result, the energy
requirements for materials and manufacturing for the remanufacturing of tires are
included in this example.
The base case considered for this study is a class 8 tractor trailer truck (gross vehicle
weight greater than 33,000 lbs. or 14,969 kg) with a fuel mileage of 5.5 mpg (30) and 18
radial tires. The life cycle inventory (LCI) for the materials and manufacturing for the
radial tires relied on available data in the literature ((31), (32), (33), (34)) and our
To appear Environmental Science and Technology, accepted April 2011
9
estimation method (4). We estimate the materials production and manufacturing for a
new 55 kg radial tire to be 3,622 + 643 = 4,265MJ (34). The estimate for the
remanufactured tire is 1365 MJ. Hence there is a 68% energy savings if only these two
phases of the life cycle are considered.
To estimate the use phase, we assume that the use phase energy of a tire is equivalent to
the fraction of the fuel required to overcome the rolling resistance divided by the number
of tires. Rolling resistance as a fraction of total fuel consumed for trucks however,
depends on many factors including driving and roadway conditions, speed, tire pressure,
tire wear and more. As a consequence, values given in the literature for the fraction of
fuel required to overcome rolling resistance vary enormously from 13% up to 47% of the
total fuel used (34). The US Department of Energy (DOE) however suggests a smaller
range from 13% to 33% (35). In order to manage this variation, we identify a midpoint
fraction (24%) with a specific measured rolling resistance coefficient of 0.0068 and then
make comparisons to this reference case. We do this because rolling resistance
coefficients can be directly measured in the laboratory under highly controlled
conditions. The key assumption is that changes in the fuel required to overcome tire
rolling resistance are proportional to the coefficient of rolling resistance (34). To make
comparisons with other tire technologies then, we use the following values of the
coefficient of rolling resistance; for conventional bias ply tires 0.0097, for new improved
radials (sometimes called low rolling resistance tires) 0.0061 and for new single-wide
tires 0.0054 (33), (35). New single-wide tires are now offered by a number of tire
companies. They can replace a pair of conventional tires when mounted together on the
axle.
An additional complication for tire remanufacturing is that because these operations can
take place at many small companies, there can be significant variation in the quality of
the retreading job. While it is true that a tire retreading operation can restore a tire to
near original performance, from the available data there is evidence that retreading can
sometimes fail to achieve like-new product performance. For example, measurements by
To appear Environmental Science and Technology, accepted April 2011
10
Michelin show that the rolling resistance for retreaded radial tires can increase between
7% to 9% compared to new radials (36).
Putting this all together requires a series of assumptions often for variables that can have
a large range of values. We tried to select values that represented central tendencies, or
to slightly bias the calculation in favor of remanufacturing. For example, our assumption
that both the new radials and the retreaded radials have the same mileage lifetime of
100,000 miles favors retreading.
For the overall use phase calculation, we assume 100,000 miles traveled at 5.5 mpg with
24% of the fuel used to overcome rolling resistance; this gives an energy value per tire of
35,640 MJ (34). If the retreaded tire has an 8% increase in rolling resistance, this adds an
additional 2851MJ for the use phase of the remanufactured tire. Now if we compare this
to the savings from the difference in the materials production and manufacturing phases
(4265 – 1365 = 2900 MJ) we see a potential savings of 49MJ for the retreading option.
But this is only about 0.1% of the life cycle energy for the new tire. This difference is
clearly within the margin of error for the life cycle energy methodology. There is no
measurable increase, nor decrease, in the total energy consumed between the two options.
If the lifetime of the retreaded tire is less, then more than one retreaded tire will be
needed and this will favor buying new. If we assume the rolling resistance fraction is
larger, say 33% instead of 24% (less starting and stopping, driving continuously at a
slightly reduced speed to decrease aerodynamic drag) then this will favor buying new. If
one can show that the performance of the retreaded tire is equal to the new tire then
retreading can produce a maximum savings of 2900 MJ (about 7.6% reduction in life
cycle energy compared to the new tire – also probably within the margin of error for the
methodology). But this would be the exceptional case, not the rule. Using the
coefficients of rolling resistance given earlier, one can calculate that choosing a retreaded
radial ply (0.0068) instead of a retreaded bias ply tire (0.0097) will save 15,199 MJ.
(This is about 28% of the bias ply tire lifecycle energy, and clearly significant). In this
calculation we assumed that the material and manufacturing energy for the bias tire was
To appear Environmental Science and Technology, accepted April 2011
11
the same as that for the radial. Other significant energy savings can be calculated by
using the new lower rolling resistance tires listed above.
4.3 25 Case Studies
The results from the two previous cases, though quite different in details, lead to rather
similar conclusions. In both cases the life cycle energy is dominated by the use phase,
and in both cases no clear answer can be given to the simple question, does
remanufacturing save energy? The answer is nuanced and depends upon many details.
When we opened this study up to still more products, we found this situation occurred
quite often. In fact, the answer to the question, does remanufacturing save energy? is
conditional and highly dependent upon current product development trends.
Furthermore, when there was a clear answer, it was just as likely that the answer was
“no” as it was “yes”.
The details for these case studies are given in Table 1, with relevant product and scenario
data, literature references, and a reference number system (1 – 25) that is carried through
to the graphical representations of the results in Figures 2 and 3 (extra literature
references (labeled ‘SI’) are provided in the Supporting Information). Figure 2 is a log-
log plot of the absolute values for the life cycle energy for the new (Y-axis) and
remanufactured products (X-axis). Figure 3 is the percent energy savings for
remanufacturing relative to the new product option in order 1 through 25.
Figure 3 clearly reveals that the answers to our question are split; there is a group of
products that can provide large relative energy savings (products numbered 1 – 8), and
there is a group of products that strongly favor buying new (products numbered 20 – 25)
and then there is a group in the middle that are more nuanced (products number 9 – 19).
The products in the first group (1- 8) include office furniture (2 cases), clothing (2 cases)
and computer equipment (4 cases). They all save energy when remanufactured, resold or
upgraded because there have been insignificant changes in the use phase energy over the
To appear Environmental Science and Technology, accepted April 2011
12
time period considered. For the office furniture there is no use phase energy. For the
computer equipment, energy efficiency improvements within the same kind of devices
over the time period (2001 – 2005) are not large enough to overcome the manufacturing
phase savings achieved by reusing. Similarly, (though not included in this study) the
refurbishing of returned new products would fall into this category.
At the other end of the figure, products 20-25 are cases where the use phase energy has
changed significantly due to efficiency mandates and/or the introduction of new efficient
technologies. Case 20 compares a remanufactured 1998 dishwasher with a much more
energy efficient 2008, Case 21 compares a CRT to an LCD display, Case 22 compares a
retreaded bias ply truck tire to a new radial truck tire, Case 23 compares a 1994
refrigerator to a 2008 model, Case 24 compares a used desktop computer to a new laptop,
and Case 25 compares a rebuilt top loader clothes washing machine to a new front loader.
In each case, choosing the remanufactured product over a new will result in a significant
additional energy requirement as indicated in Figure 3.
In the middle of this figure are a number of products (9 – 19) that require more
explanation. It should first be pointed out however that all of these cases lie between
+7% and -4% of the new product energy requirement, and we are not sure the LCI
methodology can make accurate statements in this range. Nevertheless, starting with
Case 9 we compare a retreaded radial truck tire to a new radial truck tire. A savings is
indicated here because we have not included the potential loss in performance for the
retreaded tire. As discussed earlier, this loss can be substantial, with the result that the
potential savings shown would be reduced to zero.
Case 10 is for the refilling of a toner ink cartridge. The projected savings would be 6%
provided the refilled cartridge functioned as new. In this area there is very little data on
the performance of refilled cartridges and nothing we have found from the
remanufacturing industry. However one report, commissioned by HP, suggests that it
takes 101 sheets of paper to print 100 good copies with a new cartridge and 114 to print
100 with a refilled cartridge (37). If this data were correct, the embodied energy in the
To appear Environmental Science and Technology, accepted April 2011
13
extra paper and electricity needed to print the additional 13 pages would be enough to
offset the projected savings. However, in order to make all assumptions in favor of
remanufacturing, data and results presented in the tables and figures, for cartridges as
well as other products, assume that the remanufactured products perform like-new and do
not experience such degradation in performance.
Case 11 represents the remanufacturing of diesel engines. This product has been studied
by Sutherland and co-workers who indicate a large potential savings due to avoided
materials production and manufacturing (17). Furthermore the energy efficiency of
diesel trucks has been essentially flat at about 5.5 mpg over the time period 1975 to 2006
(30). Hence we have calculated a potential 5% energy savings. At the same time, it is
clear that even a small reduction in the fuel economy of a rebuilt engine or improvement
in the new could offset this gain. For example, a change of only 0.025 mpg would be
enough to undo this savings.
Cases 12 through 19 are dominated by two different sizes of electric motors. The smaller
(22 kW) comes under the EPAct regulation of 1992, while the larger (200 kW) does not.
The cases essentially compare different new motor efficiency ratings, with various
rewound motors. The key piece of information included in these calculations is that we
used the DOE recommendation to reduce the efficiency of the rewound 22 kW motors by
0.5% and for the rewound 200 kW motors by 1% (38), (39). This difference is enough to
shift the result, in terms of energy usage, in favor of buying new. Again, we state our
doubts whether the LCI methodology can really make meaningful statements when the
differences are so small.
All of these cases are plotted in terms of absolute energy requirements in Figure 2. Points
that lie above the dividing line favor remanufacturing, while those below, favor buying
new. Note that the energy resources used by motors and engines are large, and so small
performance improvements, if they can be substantiated, could represent significant
savings in magnitude. They would however, be small relative to the total energy
resources used.
To appear Environmental Science and Technology, accepted April 2011
14
5. Discussion
When taken as a whole, it seems that making general energy savings claims for
remanufacturing is not advisable. It happens that, historically, remanufacturing did save
energy (and materials too) when products were unpowered. But current design trends of
powering up products appear to have altered the energy resources usage substantially.
That is, products that used to have no use phase are now powered. For example, rakes,
snow shovels, and hammers, are now leaf blowers, snow blowers, and power tools. This
trend brings convenience and reduces human toil, but at the same time subsequent
improvements in energy efficiency could work to reduce the potential energy savings
promised by remanufacturing. (In Figure 2 this phenomena would be represented by
products moving from on, or very near to the Y-axis – where remanufacturing would
clearly save, to the dividing line – where the outcome involve small differences between
large numbers). It has often been proposed to design using a modular platform in order to
incorporate new features in used products. This could be a significant advancement for
remanufacturing. For the purpose of this paper, this would mean incorporating energy
efficiency improvements. However, it was also observed in this study that many of the
major efficiency improvements in products are not incremental but radical, with major
transformations in the product architecture, inhibiting such upgrades. Examples include
desktop to laptop computers, top-load to front-load washing machines, and bias-ply to
radial tires. On the other hand, the upgrading of components could be accomplished if
they were standardized.
At the same time, other old benefits still accrue. Remanufacturing does provide local
skilled jobs, generally reduces transportation when the primary materials come from far
away, and may displace some primary production if the remanufactured product is truly a
substitute for a new product. Concerning transportation, in the sensitivity analysis, it
became apparent that transportation could become an issue for some extreme cases such
To appear Environmental Science and Technology, accepted April 2011
15
as the air transport of new laptop and notebook computers from Asia to the United States.
This can add substantially to the energy requirements of new products. Under these
conditions, remanufacturing can appear energy saving. The case for laptops presented in
this paper (# 4 as per Table 1) shows that even if transportation is not included, the
energy savings by reusing the old laptop can be close to 50% of the manufacturing plus
use phase energy requirements of the new laptop. Adding international transport will
increase these relative energy savings to 58% making reuse even more favorable (details
are available in the Supporting Information). Amongst other case studies, results in Table
1 include transportation for textiles, toner cartridges, and furniture, while for appliances,
and engines a sensitivity test showed the transportation contribution to be negligible
(more information in S.I.). In closing, we point out that while there are many additional
aspects of remanufacturing that could be explored, one that strikes us as particularly
important is the degree to which the remanufactured products actually substitute for new.
This is a research issue unto itself. Past work indicates that the relationship can be quite
complex, and in some cases the two products can end up being more like complements
than substitutes (40) (41).
Acknowledgements
Authors Sahil Sahni and Avid Boustani have contributed equally to this study. We would
like to thank Dr. Elsa Olivetti of MIT for her contributions to this study, and we
gratefully acknowledge the support of the MIT Energy Initiative (MITEI) and the
Singapore MIT Alliance (SMA).
Supporting Information Available
Additional assumptions, references for Table 1 and Technical Reports, are available free
of charge via the Internet at http://pubs.acs.org.
References 1. Lund, R. T. The remanufacturing industry: hidden giant, Boston University, 1996. 2. Hauser, W., Lund, R. T. Remanufacturing: operating practices and strategies, perspectives of the management of remanufacturing businesses in the united states. Department of Manufacturing Engineering, Boston University, 2008. 3. Static Control Components, Inc., v. Lexmark, Inc, 615 F.Supp.2d 575 (2009).
To appear Environmental Science and Technology, accepted April 2011
16
4. Duque, N. C.; Gutowski, T.G.; Garetti, M. A. Tool to Estimate Materials and Manufacturing Energy for a Product. IEEE - International Symposium on Sustainable Systems and Technology, Washington D.C. 2010. 5. Smil, V. Energy in nature and society: general energetics of complex systems, MIT Press, 2008. 6. Ashby, M. Materials and the Environment: Eco-informed material choice., Butterworth-Heinemann, 2009. 7. Gutowski, T.G.; Branham, M.S.; Dahmus, J. B.; Jones, A; Thiriez, A; Sekulic, D.P. Thermodynamic analysis of resources used in manufacturing processes. Environmental Science and Technology, 2009, 43, 1584–1590 8. Intlekofer, K.; Bras, B.; Ferguson, M. Energy implications of product leasing, Environmental Science and Technology, 2010, 44(12), 4409-4415. 9. Kiatkittipong, W.; Wongsuchoto, P.; Meevasana, K.; Pavasant, P. When to buy new electrical/electronic products? Journal of Cleaner Production, 2008, 16, 1339-1345. 10. Neto, J. Q. F., Bloemhof, J. M. The environmental gains of remanufacturing: evidence from the computer and mobile industry, ERIM Report series research in management, 2009. ERS-2009-024-LIS. 11. Kim H.C.; Keoleian G.A.; Horie, Y.A. Optimal household refrigerator replacement policy for life cycle energy, greenhouse gas and cost., Energy Policy, 2006, 34 (15), 2310-2323. 12. Geyer, R.; Van Wassenhove, L.; Atalay, A. The Economics of Remanufacturing under limited Component Durability and Finite Product Life Cycles. Management Science. 2007, 53 (1), 88-100. 13. Kerr, W.; Ryan, C. Eco-efficiency gains from remanufacturing: A case study of photocopier remanufacturing at Fuji Xerox Australia. Journal of cleaner production, 2001, 9(1) 75-81. 14. Ijomah, W. L.; Hammond, G.P.; Childe, S. J.; McMahon, C. A. A robust description and tool for remanufacturing: A resource and energy recovery strategy. EcoDesign 2005, Proceedings of the Fourth International Symposium on Environmentally Conscious Design and Inverse Manufacturing. Tokyo, 2005. 15. Steinhilper, R.; Hieber, M. Remanufacturing – the key solution for transforming “downcycling” into “upcycling” of electronics. Proceedings of the IEEE International Symposium on Electronics and the Environment, Denver, 2001. 16. Nasr, N.; Thurston, M. Remanufacturing: A key enabler to sustainable product systems. Proceedings of the 13th CIRP International Conference on Life Cycle Engineering, Lueven, 2006. 17. Sutherland, J. W.; Adler D. P.; Haapala, K. R.; Kumar, V. A comparison of manufacturing and remanufacturing energy intensities with application to diesel engine production, CIRP Annals - Manufacturing Technology, 2008, 57, 5-8. 18. Smith, V. M.; Keoleian, G. A. The value of remanufactured engines: life cycle environmental and economic perspectives, Journal of Industrial Ecology, 2004, 8 (1-2), 193-221. 19. Ferrer, G. The economics of tire remanufacturing, Resource, Conservation, and Recycling, 1997, 19 (4), 221-225.
To appear Environmental Science and Technology, accepted April 2011
17
20. Rosenfeld, A. H. The California Vision: Reducing Energy Intensity 2% Per Year. ACEEE Conference on Energy Efficiency as a Resource, 2003, http://www.aceee.org/conf/03ee/Rosenfeld-WSw.pdf. 21. Association of Home Appliance Manufacturers, Energy Efficiency and Consumption Trends, 2008. 22. Association of Home Appliance Manufacturers, Major home appliance saturation and length of first ownership, Appliance Information Center, 2001. 23. US Department of Energy, EERE. Building energy databook, 2009. 24. Truttmann, N.; Rechberger, H. Contribution to resource conservation by reuse of electrical and electronic household appliances. Resources, Conservation & Recycling, 2006, 48(3), 249-262. 25. Baldwin, S. The Future of Generation: Technologies and Fuels Efficiency and Renewables, 2002. Also see: http://www1.eere.energy.gov/vehiclesandfuels/pdfs/deer_2003/plenary/deer_2003_baldwin.pdf 26. Kemna, R.; Elburg, M. V.; Li, W.; Holsteijn, R.V. Methodology Study Eco-Design of Energy Using-Products, Van Holsteijn en Kemna BV and European Commission, 2005. 27. Chapman, P. F., Roberts, F. Metal resources and energy, Butterworth's monograph in materials, 1983. 28. Dahmus, J. B.; Gutowski, T.G. Can Efficiency Improvements Reduce Resource Consumption? A Historical Analysis of Ten Activities, Journal of Industrial Ecology, accepted for publication, 2011. 29. Rubber Manufacturers Association, Tire industry facts, Fact Book 2002: US tire shipment activity report for statistical year, 2001. 30. Davis, S. C.; Diegel, S. W; Boundy, R.G. Transportation energy databook: edition 28, Center for transportation analysis, Oak Ridge national laboratory, 2009. ORNL-6984. 31. Rubber Manufacturers Association, Typical Materials Composition of a Truck Tire. Retrieved 07/22/2009, from http://www.rma.org/scrap_tires/scrap_tire_markets/scrap_tire_characteristics/#anchor135840. 32. Brown, H. L.; Hamel, B. B.; Hedman, B. A. Energy analysis of one hundred and eight industrial processes, by Fairmont Press, 1985. 33. Gaines, L., Stodolsky, F., Cuenca, R., Eberhardt, J. Lifecycle analysis for heavy vehicles, Centre for Transportation Research, Argonne National Laboratory, 1998. 34. Boustani, A. Remanufacturing and Energy Savings. Master of Science Thesis, Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, 2010. 35. Bradley, R. Technology Roadmap for the 21st Century Truck Program: A Government-Industry Research Partnership. 21CT-001, Office of Heavy Vehicles Technologies of US Department of Energy, Oak Ridge, Tennessee, December, 2000. http://www.osti. gov/fcvt/21stcenturytruck. pdf. 36. Personal Communication with Don Baldwin at Michelin Center of Technology, Research and Development. July 23, 2009. 37. Four Element Consulting LLC. Laserjet Cartridge Lifecycle Environmental Impact Comparison Refresh Study: Hp laserjet 10a print cartridge vs. remanufactured brands in North America, 2008.
To appear Environmental Science and Technology, accepted April 2011
18
38. Electrical Apparatus Service Association; Association of Electrical and Mechanical Trades, Rewind study and good practice guide to maintain motor efficiency. The effect of repair/rewinding on motor efficiency, 2003. 39. MotorMaster+ Version 4, http://www1.eere.energy.gov/industry/bestpractices/software_motormaster.html 40. Thomas, V. M. Demand and Dematerialization Impacts of Second-Hand Markets: Reuse or More Use? Journal of Industrial Ecology, 2003, 7, 65-78. 41. Mokhtarian, P. L. Telecommunications and Travel: The Case for Complementarity. Journal of Industrial Ecology, 2003, 6, 43-57.
To appear Environmental Science and Technology, accepted April 2011
19
Table 1: Summary for 25 Case Studies. See supporting information for further
details and references.
To appear Environmental Science and Technology, accepted April 2011
20
Figure 1: Total life cycle energy consumption of new and remanufactured
refrigerators for the decision analysis of years 1966, 1980, 1994 and 2008. The insert
shows the breakdown into different life cycle phases for the new refrigerators.
To appear Environmental Science and Technology, accepted April 2011
21
Figure 2: Life cycle energy consumption (normalized units) for new and
remanufactured versions of all 25 product-case-studies. For the reference number
refer to Table 1. Note both axes are in logarithmic scale.
To appear Environmental Science and Technology, accepted April 2011
22
Figure 3. Percentage life cycle energy savings by choosing to remanufacture over
replace with new for the 25 different product cases. The reference numbers and
normalized units can be referred from Table 1.
Table of Contents Brief:
This study shows that in some cases remanufacturing may not save energy due to use
phase efficiency improvements and design trends.